QuickAI
An all-in-one AI SaaS platform that puts text generation, image creation, and smart editing in a single dashboard
What is it?
QuickAI is a multi-tool generative AI SaaS platform I designed and built from the ground up. Instead of juggling five different AI tools across five tabs, QuickAI brings text generation, AI image creation, smart image editing, and resume analysis into one clean, unified interface — built for content creators and professionals who want AI superpowers without the friction. It runs on a freemium model with plan-based access control and a community feed for sharing creations.
What I Did
I owned the entire product — from deciding which AI APIs to integrate, to designing the subscription architecture, to building the frontend and backend. I handled API orchestration across Google Gemini, ClipDrop, and Cloudinary AI, implemented a freemium billing model with Clerk, and built the community showcase with Optimistic UI. Every feature, every integration, and every product decision was driven by me alone.
Tech Stack & Tools
- AI SaaS
- Google Gemini
- ClipDrop
- Cloudinary AI
- Clerk Auth
- MERN Stack
- Freemium
- Full-Stack
- Personal Project
AI Tool Architecture
Designed a modular tool architecture where each AI capability (writing, image gen, editing, resume review) lives as an independent module sharing a common auth and usage-tracking layer. This made it easy to add tools without breaking existing flows.
AI Writing Assistant
Integrated Google Gemini 2.0 Flash to power full-length article generation and SEO-optimized blog title suggestions. Engineered prompt templates on the backend to ensure consistent, high-quality outputs regardless of how the user phrases their input.
AI Image Studio & Smart Editing
Wired ClipDrop's API for text-to-image generation. Layered in Cloudinary AI for instant background removal and generative fill-based object removal — giving users a lightweight, browser-native image editing suite without needing Photoshop.
AI Resume Reviewer
Built a PDF resume analysis pipeline — users upload their resume, it gets parsed and sent to Google Gemini with a structured review prompt, and returns actionable feedback with clearly identified strengths and weaknesses. Handled PDF parsing and prompt engineering entirely on the server side.
Freemium Subscription Model
Implemented plan management using Clerk's user metadata to distinguish Free vs. Premium users. Usage tracking is enforced server-side — free users hit a cap per tool per day, premium users get unlimited access. No traditional payment table needed; all plan state lives in Clerk metadata.
Community Showcase with Optimistic UI
Built a social feed where users can publish AI-generated creations and like others' work. Implemented Optimistic UI — likes update instantly on the client before the server confirms, making interactions feel snappy and native rather than laggy.
- AI Tools in One Platform
- 5+
- Writing, image gen, background removal, object removal, resume review
- AI APIs Orchestrated
- 3
- Google Gemini 2.0 Flash, ClipDrop, and Cloudinary AI working in tandem
- Freemium Subscription Model
- 2-Tier
- Free vs. Premium with usage caps enforced server-side via Clerk metadata
- Community Like Feedback
- Instant
- Optimistic UI pattern — zero perceived latency on social interactions
- AI Writing Assistant — full articles & SEO blog titles via Gemini 2.0 Flash
- AI Image Studio — text-to-image generation via ClipDrop
- Background Removal — one-click via Cloudinary AI
- Object Removal — generative fill to erase unwanted elements
- AI Resume Reviewer — PDF upload → strengths, weaknesses & actionable feedback
- Freemium model — Free vs. Premium with server-side usage tracking
- Clerk authentication with metadata-driven plan management
- Community Showcase — publish creations and like others' work
- Optimistic UI on likes — instant feedback before server confirmation
- Fully responsive across all devices
Orchestrated 3 independent AI APIs (Gemini, ClipDrop, Cloudinary) under one unified backend
Freemium gating via Clerk metadata — no billing DB table, plan state managed at the auth layer
Prompt engineering on the server side for consistent Gemini outputs regardless of user input quality
Optimistic UI pattern for community likes — client state updates before server round-trip
PDF parsing pipeline feeding directly into a structured Gemini review prompt
Modular tool architecture — each AI feature is independently extensible
Ready to build something amazing?
Interested in learning more about this project or starting a new full-stack development journey? Let's talk.